We recently completed a clinical trial testing the effectiveness of a care management program within primary care for older adults with Alzheimer's disease. Based on the clinical trial, which demonstrated the effectiveness of collaborative care, we identified the need for more comprehensive care models. These future models must address the fragmentation of care across multiple sites and providers beyond the primary care setting. To provide a more accurate picture of the entirety of care across the continuum of care, we must link data across multiple providers, health care systems, and payors. As part of the clinical trial recruitment process, we implemented a comprehensive cognitive screening and diagnosis program in primary care. Thus, we now have cognitive function data on 4,197 older primary care patients. This allows us to categorize patients into three mutually exclusive categories: dementia (n=147) cognitive impairment, no dementia (n=98);or no cognitive impairment (n=3,952). We can also track the process, outcomes, and costs of care within our health care system for all of these patients through access to clinical practice data in an advanced electronic medical record. However, these clinical practice data do not include care provided in the home or in skilled nursing facilities, for example. We now seek to establish a comprehensive database that would link our clinical trial and local electronic medical record system data with data from Medicare claims, Medicaid claims, and the Minimum Data Set. The long-range future goal of our research is to help design the next generation of care management interventions for older adults with Alzheimer's disease. The current project will allow us to identify targets for quality improvement and cost savings. This progression from descriptive epidemiologic to intervention studies is a hallmark of our past research. The specific aims of this project are to: 1) build a comprehensive longitudinal dataset across the continuum of care (home, ambulatory, inpatient, nursing home) for a large sample of vulnerable elders;2) describe and compare the frequency and timing of transitions in care among older adults with dementia, cognitive impairment-no dementia, or normal;3) describe and compare the costs of care across the continuum of care among older adults with dementia, cognitive impairment-no dementia, or normal;and 4) compare the findings from this cohort of urban poor older adults with findings from the nationally representative Health and Retirement Study (HRS). PUBLIC HEALTH RELEVANCE: Over 4.5 million Americans suffer from Alzheimer's disease. Federal, state, and personal finances are already strained to provide care for the current population, yet the number of older adults with Alzheimer's disease is expected to triple by 2050. Our research seeks to describe the costs and outcomes of care across the full continuum of health care for older adults with dementia. This descriptive data will help inform new treatment models.